Alternative Forecasting Models for Farm Wheel Tractor Horsepower Purchases
Olorunnipa, Zacch Isenewa
This item is only available for download by members of the University of Illinois community. Students, faculty, and staff at the U of I may log in with your NetID and password to view the item. If you are trying to access an Illinois-restricted dissertation or thesis, you can request a copy through your library's Inter-Library Loan office or purchase a copy directly from ProQuest.
Permalink
https://hdl.handle.net/2142/69882
Description
Title
Alternative Forecasting Models for Farm Wheel Tractor Horsepower Purchases
Author(s)
Olorunnipa, Zacch Isenewa
Issue Date
1987
Department of Study
Agricultural Economics
Discipline
Agricultural Economics
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
Ph.D.
Degree Level
Dissertation
Keyword(s)
Economics, Agricultural
Abstract
Previous research efforts on tractor demand have concentrated mainly on hypothesis testing and providing explanations for underlying intrinsic relationships among variables. Very little attention has been devoted to forecasting tractor demand.
The primary objective of this study was to forecast U.S. farm wheel tractor horsepower purchases using alternative forecasting techniques. The methodological approach adopted involved the specification and estimation of three time series models and two econometric models. The time series models included a univariate ARIMA model and two multivariate models. The latter were comprised of an unrestricted (or Sim's) vector autoregression (SVAR) and a restricted (or Bayesian) autoregression model (BVAR). Correspondingly, the econometric models consisted of a single equation and a multiple equation tractor demand models. Each model was estimated over the period 1963 quarter three to 1983 quarter two and them used to generate six-steps ahead sequential forecasts over a ten quarter forecasting horizons (1983 quarter three to 1985 quarter four). Prior to each set of forecasts the models were updated with new market information.
The results indicate that with respect to the percentage root mean squared error (PRMSE), the time series models strikingly out-performed the econometric models at all forecast horizons. With respect to turning point predictions all the models performed quite satisfactorily with the time series models performing slightly better than the econometric models. The relatively poor performance of the econometric models was attributed to the possibility of structural change especially during the forecasting period. In addition, it was also possible that errors were compounded in the sense that the forecasts of the exogenous variables were required in order to forecast the endogenous variables.
A comparison of the relative performance of the time series models indicated that the univariate ARIMA model gave lower PRMSE at short-term forecasts horizons (up to two quarters ahead) but was out-performed by the VAR models at longer forecasts horizons. The implication of this for model users is that ARIMA models may be more reliable for tractor horsepower sales' forecasts required for short-term decisions such as inventory control, whereas, for longer-term forecasts required for planning production, the VAR models may be the appropriate choice. (Abstract shortened with permission of author.)
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.
